A discussion which doesn’t seem to want to go away is one (probably mis)characterized as RDF versus microformats (or more broadly the “uppercase” Semantic Web versus the “lowercase” semantic web - essentially Tim Berners-lee’s ambitious project to reconstruct the web versus today’s web with richer semantics).
I must say that I am largely convinced, as much through experience as simply rhetoric, that the path of least resistance typically characterizes technology adoption, which is a fundamental motivator for the microformats project. However, it is also true that disruptive technologies, which require considerable investment in new skills for their adopters do appear and obsolete existing technologies. This has happened on the web with CSS, which despite the considerable cost of adoption, has, among professional developers at the very least, essentially entirely replaced the use of HTML for marking up web page appearance.
Will this happen with “The Semantic Web”? I suspect at the very least it will be a very long haul and big ask, based on the experience of trying to get even professional web developers to adopt slightly different HTML practices to create valid HTML and accessible sites. The Semantic Web project is particularly dependent on the network effect of machine readable data, which means it is very dependent on widespread adoption before it is of any real value at all.
The issue has recently been vigorously discussed by Ian Davis with comments by Tantek Çelik, and a and a thoughtful detailed rejoinder by Ian.
[tags]microformats, RDF, CSS[/tags]
{ 3 } Comments
For something to spread beyond the theory and academic discussions, it needs to solve a pain, and it needs to give you immediate feedback.
CSS solves a pain, enough people can agree to that. And it gives you immediate feeback, just refresh the page.
Microformats solves a pain, I got rid of all the iCard and iCal files I had to maintain before (or were just too lazy to create). And it gives me immediate feedback. I can see it in action on my site, I can see it in action on your site.
Maybe someday in the future there will be smart semantic agents crawling the Webs. And although my site is not in compliance with these agents, it’s not a pain I’m feeling right now. And since they don’t exist, I wouldn’t even be able to tell if the information I’m publishing is relevant or not.
Few people want to go through the trouble of creating data that someone someday may be could use for some purpose.
Would the semantic Web get there? If there was a Netscape Navigator or IE to drive adoption, it would rise quickly.
Until then …
Aren’t some of the best innovations solutions to problems that you don’t realize exist yet? Just because it doesn’t pain you doesn’t mean its not a problem. It just means we have grown complacent with the way things are and how badly our data will interoperate.
Hi, with regards to your last point that:
The Semantic Web project […] is very dependent on widespread adoption before it is of any real value, I think it’s worth separating out “Semantic Web Technologies” and “The Semantic Web Project”.
The technologies (eg: OWL, RDF, SPARQL) are real, used by, and useful to, developers now.
As I explain here: http://semwebdev.keithalexander.co.uk/blog/posts/semanticweb-image-tagclouds I think there’s perhaps been too much emphasis on ‘the Project’, or ‘the Vision’, or whatever, to the extent that there is a general impression that semantic web technologies (like RDF) are only of any use if you want to contribute your data to this grand overarching structure of the ‘The Semantic Web’ - which, according to various pundits, is ‘doomed’ anyway.
Actually, while Microformats require widespread adoption to have any real value, RDF doesn’t require any extra adoption at all to be useful. RDF’s flexibility as a data model can make it immediately worthwhile to the developer using it, and the ability to merge and mash-up the data with other data sources is a boon.
Post a Comment